Overclock.net › Forums › Components › Mice › somewhat objectively evaluating sensor responsiveness
New Posts  All Forums:Forum Nav:

somewhat objectively evaluating sensor responsiveness - Page 55

post #541 of 700
Just did a ADNS-6010 with this method: http://www.overclock.net/t/1563813/somewhat-objectively-evaluating-sensor-responsiveness/470#post_24495260




Tons of smoothing on that one. Can you figure out how to exactly interpret the amount of counts being reported after swiping off the surface? MLT04 instantly gives zero report, A3095 1 count before zeroing out at 800cpi and then 8 counts at 3600cpi - the A6010 like 16 counts at 400/800cpi and 20 at 1600/2000cpi.

*amount of reports, rather
Edited by HAGGARD - 2/7/16 at 7:56pm
post #542 of 700
Thread Starter 
looks like the mcu added an exponential filter. just divide the height of the 2nd point by the 1st (or 3rd by 2nd, etc..) to figure out the constant.

wth i must have missed that post. that's any interesting way to test the sensor that i hadn't considered thumb.gif maybe your g400 results explain the 3090 mystery?
keep in mind that at 1m/s, it takes 1ms to travel the ~1mm size of the imaged area
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
post #543 of 700
Yeah true, they get progressively more closely spaced, so that is probably an additional filter and not the sensor-level smoothing we usually encounter.

Sorry, but what's the significance of the tracking speed here; are you implying at too low speed I could be introducing noise past swiping off the surface?
post #544 of 700
Thread Starter 
it means for lower speeds, the drop isn't as sharply defined.
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
post #545 of 700
Right.

Here's something else I found: My table has wood panels and so there's cracks there. LED systems just glomp over those, but the laser can't track, so essentially I go off the surface for a brief moment and then it tracks again. But instead of recovering to actual tracking speed immediately it filters its way there too. That's the method you had proposed back when I originally made that post. Should be easy to build a setup with large enough cracks for LED to skip too, then we can maybe judge different types of smoothing.

post #546 of 700
Thread Starter 
well depending on how the sensor works, its lift-off detection might just immediately cut off the tracking. so a sharp drop doesn't always mean no smoothing
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
post #547 of 700
True. Are you aware of any mice that employ actual lift-off detection like that? The sudden tracking after the crack/skip could help (not only) with those: As opposed to a normal swipe you are not gradually accelerating here and so the smoothing algorithm should be more directly apparent. Both for the cut and the recovery, is it realistic to just count the reports and refer to that as the smoothing range, or is there framerate or something to consider too? I. e. 0 reports MLT04 = 0ms; 1 report A3095 @ 800cpi = 1ms/@ 3600cpi = 8ms; A6010 (this implementation) = 16-20ms. All at 1kHz of course. At least comparatively this should hold true, no?

This is how the MLT04 recovers:

Essentially instant cut and instant recover. I just held up a hard pad next to my desk pad and swiped it off from the former onto the latter.
Edited by HAGGARD - 2/8/16 at 5:31am
post #548 of 700
my xai dose
post #549 of 700
Quote:
Originally Posted by omar231 View Post

my xai dose
...recognize it's lifted and stop operating altogether?

I mean, I'm sure qsx knows about a functionality like that, but to me it doesn't make sense that a sensor would need to recognize that it is in lift-off at all. Either movement (which is varying amounts of light) is detected or it isn't; why should the sensor distinquish between being sat still or elevated? I'm sure there's some technical reason I'm unaware of but I haven't seen it yet I don't think. Or do modern sensors implement that to combat the "jiggle" associated with lifting and setting your mouse down at an angle?
Edited by HAGGARD - 2/8/16 at 9:54am
post #550 of 700
Quote:
Originally Posted by HAGGARD View Post

...recognize it's lifted and stop operating altogether?

I mean, I'm sure qsx knows about a functionality like that, but to me it doesn't make sense that a sensor would need to recognize that it is in lift-off at all. Either movement (which is varying amounts of light) is detected or it isn't; why should the sensor distinquish between being sat still or elevated? I'm sure there's some technical reason I'm unaware of but I haven't seen it yet I don't think. Or do modern sensors implement that to combat the "jiggle" associated with lifting and setting your mouse down at an angle?

Surface quality feature?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Mice
Overclock.net › Forums › Components › Mice › somewhat objectively evaluating sensor responsiveness